There are a lot of subsystems in Pharo, and being a bear of very little brain, I have a hard time relating Zinc, Calypso, &c &c to, well, whatever they are. I presume there is somewhere a list of topic/name/PFX triples for guidance. Can some kind soul tell me where it is?
On 13 April 2018 at 23:01, Marcus Denker <marcus.den...@inria.fr> wrote: > > > > On 13 Apr 2018, at 12:40, Sven Van Caekenberghe <s...@stfx.eu> wrote: > > > > > > > >> On 13 Apr 2018, at 12:19, Joe Shirk <j.b.sh...@gmail.com> wrote: > >> > >> I've been a lurk-fan for a long time but this brings up something that > distressed me. Richard Eng, Smalltalk Renaissance hero loves to say > Smalltalk's grammar/syntax fits on a postcard. > >> > >> But the vocabulary doesn't. There is nothing English-like about the > always expanding bewildering library namespaces. > >> > > The package names that just use the “project name” can be problematic… too > many words. e.g. “Hiedra”? No idea. (there are ideas of how to improve, I > will not list them here as this should > not turn into discussion about this issue). > > The way we present packages (and their granularity) is not “right”. > Namespaces are a problem in addition… > > So yes: we have a lot of thing to improve! > . > >> GT what? Oh a newbie might eventually figure out it means Glamorous > Toolkit. These are meaningless brands. In this drive to come up with > creative names for "just objects" that explain nothing at all, Smalltalk is > becoming like Java or PHP hell. > >> Just look at those examples through the eyes of a novice. The purity is > nowhere to be found. > >> :( > > > > You are right, but in 'the real world' it is no longer possible to > reserve the nice, simple names for just one variant. The prefixes are a > poor mans namespace mechanism. You have to read over them. > > > > Inspector, EyeInspector, GTInspector, ... > > > > I rather have cool alternatives and the development of new ideas than > 'one ring to rule them all' or no/slow progress. Remember that we develop > in a live system, changing things while testing them, this is often hard. > Alternative subsystems help a lot. > > It should be clear that what we have is what we managed to do, not what we > dreamed about… I, too, would like to have this clean, nice, small, amazing > system… but it is not always easy. > > There is a lot we can (and will!) improve! > > Marcus > > >
